• 방송공학회논문지
• /
• 제27권2호
• /
• pp.185-197
• /
• 2022

최근 객체 인식에 높은 성능을 가진 딥러닝 네트워크가 나오고 있다. 딥러닝을 이용한 객체 인식의 경우 성능 향상을 위해 학습 데이터 셋 구축이 중요하다. 데이터 셋을 구축하기 위해서는 이미지를 수집하고 라벨링 해야 한다. 이 과정은 많은 시간과 인력이 필요하다. 때문에 오픈 데이터 셋을 사용한다. 그러나 방대한 오픈 데이터 셋을 가지고 있지 않는 객체도 존재한다. 그 중 하나가 번호판 검출과 인식에 필요한 데이터이다. 이에 본 논문에서는 이미지를 최소화 하여 대용량 데이터 셋을 만들 수 있는 인조 번호판 생성기 시스템을 제안한다. 또한 인조 번호판 배치구조에 따른 검출률을 분석했다. 분석결과 가장 좋은 배치구조는 FVC_III, B이며 가장 적합한 네트워크는 D2Det이었다. 인조 데이터셋 성능은 실제 데이터셋의 성능보다 2~3%가 낮았지만, 인조 데이터를 구축하는 시간이 실제 데이터셋을 구축하는 시간보다 약 11배 빨라 시간적으로 효율적인 데이터 셋 구축 시스템임을 증명하였다.

• 한국농공학회:학술대회논문집
• /
• 한국농공학회 2001년도 학술발표회 발표논문집
• /
• pp.59-64
• /
• 2001

The objectives of the thesis are to propose a pattern classification method for remote sensing data using artificial neural network. First, we apply the error back propagation algorithm to classify the remote sensing data. In this case, the classification performance depends on a training data set. Using the training data set and the error back propagation algorithm, a layered neural network is trained such that the training pattern are classified with a specified accuracy. After training the neural network, some pixels are deleted from the original training data set if they are incorrectly classified and a new training data set is built up. Once training is complete, a testing data set is classified by using the trained neural network. The classification results of Landsat TM data show that this approach produces excellent results which are more realistic and noiseless compared with a conventional Bayesian method.

• Structural Engineering and Mechanics
• /
• 제74권4호
• /
• pp.559-565
• /
• 2020

This paper intends to progress models to accurately estimate the behavior of fresh concrete under vibration using artificial neural networks (ANNs). To this end, behavior of a full scale precast concrete mold was investigated numerically. Experimental study was carried out under vibration with the use of a computer-based data acquisition system. In this study measurements were taken at three points using two vibrators. Transducers were used to measure time-dependent lateral displacements at these points on mold while both mold is empty and full of fresh concrete. Modeling of empty and full mold was made using ANNs. Benefiting ANNs used in this study for modeling fresh concrete, mold design can be performed. For the modeling of ANNs: Experimental data were divided randomly into two parts such as training set and testing set. Training set was used for ANN's learning stage. And the remaining part was used for testing the ANNs. Finally, ANN modeling was compared with measured data. The comparisons show that the experimental data and ANN results are compatible.

• Journal of the Korean Data and Information Science Society
• /
• 제15권2호
• /
• pp.499-506
• /
• 2004

The Unscented Kalman Filter is known to outperform the Extended Kalman Filter for the nonlinear state estimation with a significance advantage that it does not require the computation of Jacobian but EKF has a competitive advantage to the UKF on the performance time. We compare both algorithms on training the artificial neural network. The validation data set is used to estimate parameters which are supposed to result in better fitting for the test data set. Experimental results are presented which indicate the performance of both algorithms.

• Journal of Advanced Marine Engineering and Technology
• /
• 제33권1호
• /
• pp.81-89
• /
• 2009

Performance tests on the 3-Dimensional Flow-Structure-Interactions-Measurement System(FSIMS) have been carried out. Experimental data obtained by the FSIMS on a floating cylinder have been used to generate a set of artificial images. Comparisons between the data obtained by the use of the artificial images and those original experimental data have been made. Another set of artificial images have also been generated based on theoretically modulated sinusoidal motions, and comparisons between the data obtained by the use of these artificial images and the theoretical ones have been carried out. It has been verified that the FSIMS has a measurement uncertainty of 0.04-0.06mm/frame for velocity vectors and 0.002-0.01mm for the cylinder's positions.

• 천문학회지
• /
• 제36권2호
• /
• pp.67-72
• /
• 2003

The basic objective of helioseismology is to determine the structure and the dynamics of the Sun by analysing the frequency spectrum of the solar oscillations. Accurate frequency measurements provide information that enables us to probe the solar interior structure and the dynamics. Therefore the frequency of the solar oscillation is the most fundamental and important information to be extracted from the solar oscillation observation. This is why many efforts have been put into the development of accurate data analysis techniques, as well as observational efforts. To test one's data analysis method, a realistic artificial data set is essential because the newly suggested method is calibrated with a set of artificial data with predetermined parameters. Therefore, unless test data sets reflect the real solar oscillation data correctly, such a calibration is likely incomplete and a unwanted systematic bias may result in. Unfortunately, however, commonly used artificial data generation algorithms insufficiently accommodate physical properties of the stochastic excitation mechanism. One of reason for this is that it is computaionally very expensive to solve the governing equation directly. In this paper we discuss the nature of solar oscillation excitation and suggest an efficient algorithm to generate the artificial solar oscillation data. We also briefly discuss how the results of this work can be applied in the future studies.

• 자동차안전학회지
• /
• 제13권4호
• /
• pp.33-38
• /
• 2021

This paper presents a comparative study between the parameter-optimized Pacejka model and artificial neural network model for the tire force estimation. The two different approaches are investigated and compared in this study. First, offline optimization is conducted based on Pacejka Magic Formula model to determine the proper parameter set for the minimization of tire force error between the model and test data set. Second, deep neural network model is used to fit the model to the tire test data set. The actual tire forces are measured using MTS Flat-Track test platform and the measurements are used as the reference tire data set. The focus of this study is on the applicability of machine learning technique to tire force estimation. It is shown via the regression results that the deep neural network model is more effective in describing the tire force than the parameter-optimized Pacejka model.

• 한국인공지능학회지
• /
• 제5권1호
• /
• pp.18-28
• /
• 2017

Recently, there are many companies that use systems based on artificial intelligence. The accuracy of artificial intelligence depends on the amount of learning data and the appropriate algorithm. However, it is not easy to obtain learning data with a large number of entity. Less data set have large generalization errors due to overfitting. In order to minimize this generalization error, this study proposed DGA which can expect relatively high accuracy even though data with a less data set is applied to machine learning based genetic algorithm to deep learning based dropout. The idea of this paper is to determine the active state of the nodes. Using Gradient about loss function, A new fitness function is defined. Proposed Algorithm DGA is supplementing stochastic inconsistency about Dropout. Also DGA solved problem by the complexity of the fitness function and expression range of the model about Genetic Algorithm As a result of experiments using MNIST data proposed algorithm accuracy is 75.3%. Using only Dropout algorithm accuracy is 41.4%. It is shown that DGA is better than using only dropout.

• 한국시뮬레이션학회논문지
• /
• 제30권3호
• /
• pp.1-9
• /
• 2021

전장에 있는 지휘관과 참모들은 상황을 인식하고 그 결과를 바탕으로 지휘결심을 통해 군사 활동을 수행하는데, 최근 정보기술의 발달과 함께 지휘결심을 지원하는 인공지능에 대한 요구가 증가하였다. 인공지능을 활용하기 위해서는 강화학습에 필요한 학습 data set의 식별, 수집 그리고 전처리가 필수적이다. 그러나 전술 C4I 체계에 저장된 적 data는 정확성, 적시성, 충분성 측면에서 인공지능 학습 data로 사용하기에 적절하지 않기 때문에 학습 data를 수집하고 훈련 시킬 수 있는 대안이 필요하다. 본 논문에서는 육군의 워게임 훈련 모델인 '창조 21 모델 훈련 data'를 활용하여 인공지능을 학습시키는 방법론을 제시하였다. 연구 범위는 군사결심수립과정과 연계하여 인공지능의 역할과 범위를 구체화하고, 그 역할에 맞추어 인공지능을 훈련 시키기 위해 창조 21 모델 연습 data를 활용하는 모델을 제시하였다. 공개가 제한되는 군사자료의 특성을 고려하여 가상의 sample data를 제작하였고, 공개가 제한되는 대한민국 육군의 교리는 인터넷에서 수집 가능한 미군 교리를 활용하였다.

• Environmental Engineering Research
• /
• 제14권2호
• /
• pp.102-110
• /
• 2009

Application of artificial intelligence (AI) approaches in eco-environmental modeling has gradually increased for the last decade. Comprehensive understanding and evaluation on the applicability of this approach to eco-environmental modeling are needed. In this study, we reviewed the previous studies that used AI-techniques in eco-environmental modeling. Decision Tree (DT) and Artificial Neural Network (ANN) were found to be major AI algorithms preferred by researchers in ecological and environmental modeling areas. When the effect of the size of training data on model prediction accuracy was explored using the data from the previous studies, the prediction accuracy and the size of training data showed nonlinear correlation, which was best-described by hyperbolic saturation function among the tested nonlinear functions including power and logarithmic functions. The hyperbolic saturation equations were proposed to be used as a guideline for optimizing the size of training data set, which is critically important in designing the field experiments required for training AI-based eco-environmental modeling.